10e021

7/31/2019 10e021

1/22

RIETI Discussion Paper Series 10-E-021

Empirical Study of IPTV Diffusion:

Comparison between Japan and Korea

KINUGASA SatoshiBank of Japan

MOTOHASHI Kazuyuki

RIETI

Yeong-Wha SAWNG

Electronic and Telecommunications Research Institute

TERADA Shinichiro

the University of Tokyo

The Research Institute of Economy, Trade and Industry

http://www.rieti.go.jp/en/
http://www.rieti.go.jp/en/http://www.rieti.go.jp/en/http://www.rieti.go.jp/en/

7/31/2019 10e021

2/22

RIETI Discussion Paper Series 10-E-021

April 2010

Empirical Study of IPTV Diffusion:

Comparison between Japan and Korea

Satoshi KINUGASA (Bank of Japan), Kazuyuki MOTOHASHI (the University of

Tokyo and RIETI), Yeong-Wha SAWNG (ETRI), and Shin-ichiro TERADA (the

University of Tokyo)

Abstract

This study analyzes a diffusion pattern of IPTV (Internet Protocol TV), based on TAM (Technology

Acceptance Model). RIETI conducted a household survey of IPTV to understand factors behind

advancement of "digital conversion" of Internet and broadcasting services in Japan.

The results of this survey are compared with those of a similar survey in Korea, conducted by ETRI

(Electronics and Telecommunications Research Institute). Our structural equation model (SEM)

analysis reveals that "ease-of-use" is an important factor in encouraging non IPTV users in both

Japan and Korea to adopt it, but that the path from "usefulness" to "adoption" is stronger for

Japanese households. Also in relation to IPTV users intention to use IPTV, Japanese were directly

influenced by usefulness, while Koreans were not.

Keywords: IPTV, diffusion of innovation, Technology Acceptance Model, Structural Equation Model

JEL Classification: L86, O33

RIETI Discussion Papers Series aims at widely disseminating research results in the form of professional papers,

thereby stimulating lively discussion. The views expressed in the papers are solely those of the author(s), and do not

present those of the Research Institute of Economy, Trade and Industry.

7/31/2019 10e021

3/22

1

1. IntroductionInternet Protocol Television (IPTV), the convergence services of television and Internet,

is being rapidly developed around the world. Many digital media telecommunications service

providers around the world are moving towards triple play offerings by combining voice, video

and high speed data services. IP-based technology is being used by the majority of service

providers to support such video services, commonly known as IPTV services. Diffusion of new

technologies results from a series of individual decisions to start using the new technology, which

is often the result of a comparison between the uncertain benefits of it and the uncertain cost of

adopting it.

Understanding the factors affecting the choice of users is essential both for researchers

studying the diffusion factors and for the manufacturers of this newborn industry. The users

motivation to use IPTV can be categorized as intrinsic factors such as seeking high quality,

content-rich, and value added services; and extrinsic factors, which include highly interactive

services and interoperable applications with other devices and platforms. IPTV now faces a new

set of challenges. Their top priority is to deliver an enhanced experience to audience. IPTV

services must meet and exceed quality levels currently offered by traditional TV players. So,

providers of IPTV must use the findings from the research to identify factors with favorable

characteristics for adoption of the IPTV services.

A history of ITPV service in Japan is not short. The first IPTV service in Japan, BBTV,

started in 2003, by Yahoo! Japan. Now IPTV services are provided by internet providers such as

KDDI and NTT Plala and CATV providers such as J:COM. However, the diffusion rate of IPTV

in Japan is only 2.5%, which is significantly lower than the CATV diffusion rate of about 44.0%

in the end of March, 2009 (Ministry of Internal Affairs and Communications, 2009). On the other

hand, the diffusion rate of IPTV in Korea is about 17%, even though its service started in July

2006, more than 3 years after Japan. Major IPTV providers in Korea are also internet providers

such as KT and SK, which are telecommunication companies such as NTT and KDDI. In addition,

the regulatory framework concerning broadcasting and telecommunication services is also similar

between both countries.

7/31/2019 10e021

4/22

2

In this paper, the difference in IPTV diffusion rate between Japan and Korea is analyzed

by estimating structural equation model (SEM), based on Technology Acceptance Model (TAM)

in the literature of innovation diffusion (Davis, 1989; Rogers, 1995). We use the results of

household surveys in Japan, conducted by RIETI and in Korea, by ETRI (Electronics and

Telecommunications Research Institute). Our structural equation model (SEM) analysis reveals

that "Ease-of-Use" is an important factor for non IPTV users to use it, and that the path from

"usefulness" to "adaption" is stronger for Japanese households, and that IPTV users intention to

use IPTV was influenced by usefulness directly in Japan, which was not influenced in Korea.

The organization of this paper is as follows. The next section describes the methodology

and data in this paper. Then, the results of SEM analysis are provides. We compared the

coefficients of SEM by IPTV users and non users, as well as Japanese users and Korean users.

The section 4 presents some discussion about the results. And, this paper is concluded with

summary of findings and some lines of future research.

2. Methodology2.1 Research Model

The objective of this study is to empirically investigate the process of diffusion of

converged media services by looking at the case of IPTV comparing between Japan and Korea.

We designed a conceptual research model by a TAM-based adoption-diffusion model. It asserts

that the intention to use the product, IPTV, is a function of three factors; Perceived Ease-of-Use,

Perceived Usefulness, and Attitude Toward the Product, IPTV.

The research model consists in applying a TAM-based adoption-diffusion model to IPTV

services so as to identify a comprehensive range of factors influencing their diffusion process. The

variables selected for the research model are wide-ranging and include major TAM variables. The

adoption-diffusion model used in this study is a structural model having Perceived Ease-of-Use,

Perceived Usefulness, and Attitude Toward the Product as mediating variables. Intention to Use

the Product was selected as outcome variables for acceptance-diffusion model.

7/31/2019 10e021

5/22

3

2.2 Hypothesis

TAM and other adoption-centered theories argue that the diffusion of an innovation

within a social network begins when an individual member comes into the awareness of it and

communicates his/her knowledge to others within the network. According to this postulate, for an

effective diffusion of a new product, a marketer must promote the product in such a way as to

create as many early adopters as possible, so that these people would, in turn, convert more

consumers into adopters through inter-consumer communication. Factors believed to influence

consumers adoption of innovations vary depending on the researcher, even though many share

basic premises on the diffusion process, which have been formulated by influential

adoption-diffusion theories such as the theory of diffusion of innovations of Rogers (1995), TRA

(Theory of Reasoned Action) by Fishbein & Ajzen (1975), TPB (Theory of Planned Behavior) by

Ajzen (1991) and TAM (Technology Acceptance Model) by Davis (1989). Rogers (1995) saw the

diffusion of innovations as a progressive process in which the adoption of an innovation, initially

only by a small number of people, gradually increases to eventually lead to a mass-market take-up.

In other words, Rogers believed that the early phase of innovation adoption was closely linked to

the personal innovative tendencies of early users.

Among early adoption-diffusion studies, Rogers & Shoemaker (1971) proposed five

factors that influence the rate of adoption and diffusion of a new technology product: relative

advantage, compatibility, simplicity, observability and trialability. According to Rogers &

Sheomaker (1971), the driving factor of the diffusion process is the relative advantage presented

by a new technology product, over the existing product, and that the more the innovative product

corresponds to consumers desire, beliefs, values and personal experience, the greater the extent

and speed of its diffusion. They further claimed that the simpler an innovative product, the faster

the rate of diffusion, as the Ease-of-Use of a new technology facilitates its broad take-up and, in

some cases, even trumps price considerations. An eye-catching new product, they also maintained,

has a better chance of being quickly adopted by the mass-market, and odds for market success are

greater for products that can be tried in advance of purchase, without financial risk to consumers.

Robertson & Gatignon (1986), in a study on the adoption of innovative cutting-edge products

7/31/2019 10e021

6/22

4

among industrial buyers, advanced the view that a competitive supplier-side environment

influenced the demand-side competition environment, thereby, also influencing the diffusion of

new technology products. Here, the competitive environment has as its key components,

industrial competitiveness, company reputation, technology standardization and vertical inter-firm

cooperation. In TAM, a dominant paradigm in todays adoption-diffusion research, attitude and

behavioral intention-related variables are believed to influence individuals actual adoption of

innovations, through the intermediary of Perceived Ease-of-Use and Perceived Usefulness (Davis

et al., 1989). Lee et al. (2002) report, in their study on the banking industrys adoption of new

technology products, that communication was an important predictor of the actual acceptance of

new products. In other words, Perceived Ease-of-Use and Perceived Usefulness are determinants

of consumers behavioral intention concerning technology acceptance, and the influence of

external variables on technology acceptance is mediated by Perceived Ease-of-Use and usefulness

(Agarwal and Karahanna, 2000). Joo and Kim (2004), meanwhile, found, in an investigation of

technology acceptance in the internet market, that innovativeness, external environment and

organizational characteristics were the three most important determinants of acceptance.

In this study, we draw on the adoption-diffusion model using four key influence factors

for technology acceptance in TAM, namely Perceived Ease-of-Use (PEoU), Perceived

Usefulness(PU), Attitude Toward the Product(ATP), and intention to use the product(IUP). And

other variables also used including simplicity, similarity, complexity, trialability, QoS, compativity,

substitution, relative advantage, complimentarity, and satisfaction, as influence factors. Using

these variables, we formulated the following hypotheses on the adoption-diffusion of IPTV.

H1: PEoU was positively related with PU.

H2: PU was positively related with ATP.

H3: PEoU was positively related with ATP.

H4: ATP was positively related with IUP.

H5: PU was positively related with IUP.

7/31/2019 10e021

7/22

5

2.3 Data Collection

Empirical data were collected from both users and non-users of IPTV. We designed two

different questionnaires for households currently using an IPTV service and households not

currently using one.

In Korea, the data were collected through a face-to-face interview conducted at the

homes or workplaces of the respondents. The respondents were selected among households

residing in four major Korean cities, namely, Seoul, Incheon, Busan and Daegu. 500 total copies

of the non-user analysis survey questionnaire were distributed to households that are not currently

using IPTV, which broke down by region to 200 copies to Seoul-based households and 100 copies

each in Incheon, Busan and Daegu-based households. As for the user analysis questionnaire,

designed for households that are currently using IPTV, 250 copies distributed 100 copies in

Seoul and 50 copies each in Busan, Incheon and Daegu. The survey was conducted over a period

of one month, between October 15, 2008 and November 14, 2008.

In Japan, on the other hand, the data were collected though a web questionnaire. The

respondents were selected among individuals residing all cities in Japan. 1000 total copies of the

non-user analysis survey questionnaire were distributed to households that are not currently using

IPTV. As for the user analysis questionnaire, designed for households that are currently using

IPTV, 1000 copies distributed. The survey was conducted over a period of three weeks, between

September 1, 2009 and September 20, 2009. Table 1 compares main features of survey for both

countries, and Table 2 shows some characteristics of households in our datasets.

Table 1 Characteristics of Sample

Survey Regions all the city in Japan

Survey Conductor RIETI

Surver Method web questionnaire

Survey Period 2009.09.01-09.20

Sample No. N=2000(users=1000, non users=1000)

cf. all households also use CATV without 500 of non users

Survey Regions Seoul, Incheon, Busan and Daegu

Survey Conductor ETRI

Surver Method Face to face

Survey Period 2008.10.15-11.14

Sample No. N=750(users=250, non users=500)

Seoul: 300, Busan, Daegu and Incheon: 150

cf. all households also use CATV

Japan

Korea

7/31/2019 10e021

8/22

6

Table 2 Description of the subjects

N % N %

2000 100 750 100

Male 1430 71.5 379 50.5

Femal 570 28.5 371 49.5

20s 270 13.5 146 19.5

30s 633 31.7 287 38.3

40s 640 32 213 28.4

50s 334 16.7

60s 106 5.3

over 70s 17 0.9

self-employed 253 12.7 143 19.1

Sales 209 10.5 80 10.7

Production 64 3.2 23 3.1

Clerical employee 380 19 252 33.6Technical employee 417 20.9 72 9.6

Administration 163 8.2 24 3.2

Professionals 148 7.4 57 7.6

Agriculture, fisheries,

and stockbreeder9 0.5 4 0.5

Housewives 113 5.7 69 9.2

Students 35 1.8 15 2.0

Inoccupation 151 7.6 10 1.3

Others 58 2.9 1 0.1

Junior High school 25 1.3 8 1.1

Senior High school 617 30.9 260 34.7Under Graduate 33 1.7 31 4.1

Graduate 1176 58.8 407 54.3

post graduate 149 7.5 44 5.9

0 to 2 million yen 65 3.3











over 12 million yen 177 8.9

NA 76 3.8

do not wish to disclose 121 6.1

0 to 2.5 million won 151 20.1

2.5 to 5.0 million won 436 58.1

5.0 to 7.5 million won 117 15.6

over 7.5 million won 46 6.1

Japan Korea

Total

Gender

Ages

104 13.9

Occupation

Graduation

Annual Incomes

7/31/2019 10e021

9/22

7

2.4 Measurement

This study considers variables that are common to the adoption-diffusion model and

those that are specific to each of them. Variables specific to the adoption-diffusion model

considered in this study include compatibility, observability, similarity, trialability, household

innovativeness, communication, complexity, relative advantage, perceived risk and service quality

Perceived Ease-of-Use and Perceived Usefulness.

Trialability was measured through the limited period of use, capabilities to use the

functions of a product, ability to use the product when needed and performance enhancement

resulting from the use of the product (Rogers, 2003). Quality of service was measured by the

accuracy, relevance, completeness and comprehensiveness of information provided through a

product/service, and the variety of information made accessible by it (Parasuraman et al., 1985).

Compatibility refers to the extent to which a new product corresponds to the desire, trust, values

and the past experience of a consumer (Rogers, 1995).

Household innovativeness refers to how willing a household is to adopt an innovation

(Gatignon and Robertson, 1985), in other words, how rapid a households adoption of an

innovation is, and was measured by curiosity/creativity, risk preference, voluntary simplicity,

creative re-use, and multiple use potential (Price & Ridgway, 1983).

As for functional similarity, we took into account the extent to which a consumer, based

on his/her personal experience, perceives a product/service as functionally similar to existing ones

(Martin & Stewart, 2001). Items used to measure complexity were difficulty of use, complexity of

manipulation, understanding of advanced functions and need for explanations on advanced

functions (Rogers, 1983).

Substitution effect, corresponding to the relationship between two products/services in

which the use of one dispenses the use of the other - in other words, one replaces the other was

measured with regard to TV, internet, mobile phone, DMB and other existing media, in terms of

IPTVs potential to substitute their functionalities, the content they provide and the time spent

using these media (Li, 2004). Relative advantage was measured by the comparative price

advantage of a product/service, ease and speed of manipulation and portability (Moore &

7/31/2019 10e021

10/22

8

Benbasat, 1991). Complementarity, which describes the relationship between two

products/services in which they help each other maximize their usefulness, was measured with

regard to the ability to increase the existing medias entertainment potential and usefulness as an

information access and communication tool (Jeffers & Atkin, 1996).

Perceived Usefulness was measured through the ability to quickly access information

useful to a user, usefulness of a product for conducting purchases and the variety of information

made accessible through the product (Davis, 1989).

Perceived Ease-of-Use was measured through ease of using desired functions and

convenience of methods for accessing the internet and other manipulations (Davis, 1989;

Venkatesh, 2000).

3. Analysis and Results3.1 Reliability Analysis

The dependent variables, Perceived Usefulness and Perceived Ease-of-Use, were

measured through the technology acceptance instrument developed and validated by Davis (1989).

Each variable consists of 7-point Likert scale items with 1 indicating strongly disagree and 7

indicating strongly agree.

In order to testify the reliability and validity of the variables used in TAM, we

conducted Cronbach's alpha test for each factor and performed a confirmatory factor analysis. The

results of all constructs measured in Table 3 are presented indicating sufficient level of reliability

and validity on all measures. The internal consistency reliability analysis resulted in a Cronbach's

alpha of .90 or greater for all of the factors. The reliability levels of both PU and PEoU were also

over .90 (Nunnally, 1978). Table 3 below presents the measurements of discriminant validity. It

revealed that the shared variance among variables was less than the variances extracted by the

constructs, the value on the diagonals. This showed that constructs are empirically distinct. In

conclusion, the test of the measurement model, including convergent and discriminant validity

measures, was satisfactory.

7/31/2019 10e021

11/22

9

Table 3. Reliability of Measure and Discriminant Validity through confirmatory factor analysis

Simpl. Comp. Simil. Trial. QoS Comp.Innov.

Coms.

Innov.

Info.Substit.

Relat.

Adva.Compl. Satisf.

(PEoU1) 0.912(PEoU2) 0.909(PEoU3) 0.923

(PEoU4) 0.890(PEoU5) 0.936(PEoU6) 0.883(PEoU7) 0.459(PEoU8) 0.645(PEoU9) 0.906(PU1) 0.872(PU2) 0.846(PU3) 0.948(PU4) 0.921(PU5) 0.875(PU6) 0.935(PU7) 0.917(PU8) 0.813(PU9) 0.923(PU10) 0.958

(PU11) 0.931(PI1) 0.835(PI2) 0.940(PI3) 0.945(PI4) 0.793(PI5) 0.792(PI6) 0.937(PI7) 0.944(ATP1) 0.729(ATP2) 0.874(ATP3) 0.903(ATP4) 0.562(ATP5) 0.634(ATP6) 0.745(ATP7) 0.794(ATP8) 0.901

(ATP9) 0.879(ATP10) 0.790(ATP11) 0.913(ATP12) 0.899(ATP13) 0.931(ATP14) 0.892(ATP15) 0.913(ATP16) 0.923(IUP2) 0.865(IUP3) 0.841(IUP4) 0.897(IUP5) 0.934(IUP6) 0.943

Japan 0.999 0.999 0.970 0.998 0.998 0.998 0.995 0.995 0.997 0.996 0.998 0.997

Korea 0.998 0.997 0.980 0.997 0.996 0.995 0.994 0.997 0.991 0.983 0.998 0.993

Simpl. Comp. Simil. Trial. QoS Comp.Innov.

Coms.

Innov.

Info. Substit.Relat.

Adva. Compl. Satisf.

Simplicity - -0.206 0.563 0.606 0.586

Complexity - 0.130

Similarity -

Trialability - 0.685 0.501

QoS - 0.654

Compativity -

Innov. Comsumption - 0.668 0.190 0.276 0.268 0.300

Innov. Information - 0.154 0.227 0.196 0.226

Substitution - 0.617 0.648 0.630

Relative Advantage - 0.608 0.731

Complementarity - 0.717

Satisfaction -

I want to use more in the future

Cronbach's Alpha

Covariance between Factors

Vital product and service

An important element of family life

Required service for life

Very convenient service

Very useful service

Recommend to friends and family

Bring efficiency into daily life

For CS

As family fun

As family interests

As famili leisure

The amount of information collected

The quality of information collected

Case of exchange views with someonCase of convey any info to anyone

Become an important tool in the hom

Affect the daily lives of famil

Need to be precious for family

For CATV

Well with daily leisure habits

Well with the lifestyle

Daily life in harmony with IPTVBuy new motivation is high

Always want to buy a new product

Want to buy a new product

Collect new product info before anyo

Rich in new ideas

Sensitive to new information

Sensitive to the performance

For traditional TV

Variety of information available

Requiring detailed explan to use

IPTV vs traditional TV

IPTV vs CATV

IPTV vs CS

Quite often uninterrupted

Fast Download and Upload

Very Stable Broadcasting

Good TV reception sensitivity

Accurate information is provided

Information related to desired

Information is comprehensive

Difficult to understand details

Very functional and easy to use

Ease to use without help

Generally very easy to use

Complex to use

*omitted values on covariance matrix if the value is smaller than .10.

7/31/2019 10e021

12/22

10

Figure 1. Concept model extending the classical TAM with Household Innovativeness

Figure 1 shows an overview of the concept model used for this research. The ellipse in

Figure 1 means latent variable, the rectangle shows observed variable, and the small circle shows

error variable. Single arrows means beta of regression and factor analysis, and double arrow

means covariance between the two parameters. The model proposed includes an extension of the

traditional TAM constructs (i.e. PEoU, PU, ATP, and IUP) with Household Innovativeness of

7/31/2019 10e021

13/22

11

Consumption and Information. In this study we used SEM) to assess the impact of the PEoU, PU,

ATP on customers IUP IPTV in Japan and Korea.

3.2 Measurement of model

We performed a structural analysis of the adoption-diffusion model predicting the

diffusion of IPTV, using AMOS v.18.0. The goodness of fit of the overall model proved to be

highly adequate. As shown in Table 4, the fitness measures were almost all within acceptable

range. The results of indicators in Japan, AGFI, CFI and RMSEA, are all significantly to their

recommended constructs, while some of them in Korea are little out of recommended values in

the literature. Consequently, all the measures indicated that the model fit the data.

The overall fit indices of the structural model were within the range that scholars

generally recommend. In both countries cases, we found that all of the influence between

observable variance and group factor, Trialability, QoS, Substitution, Complementarity, for

example, are 1% significant without Complexity. Hence,

Table 4. Fit indices for the measurement model

JPN KOR JPN KOR JPN KOR JPN KOR JPN KOR JPN KOR JPN KOR

Us er 3189.5 1952.3 1070 1070 2 .9 81 1 .8 25 0 .8 58 0.716 0 .9 54 0 .9 17 0 .0 45 0.058 3499 2262

Non User 2850.0 2366.3 1070 1070 2 .6 64 2 .2 12 0 .8 79 0 .8 09 0 .9 64 0.896 0 .0 4 1 0 .0 4 9 3160 2676

Recommended

Suggested by

AICCMIN df CMIN/df AGFI CFI RMSEA

(Bagozzi and Yi,

1989)

0.80 >0.90

7/31/2019 10e021

14/22

12

some measurement AGFI of Users, CFI of Non User, and RMSEA of Users were out of the

range recommended. This is partly because of the number of samples; Korean survey collected

less half size of sample than Japanese survey. In practice, however, these values are considered to

indicate a good fit (A. Seyal et.al, 2002).

The structural analysis of the adoption-diffusion model on IPTV was based on TAM. The

results of the structural analysis of the adoption-diffusion models on IPTV are showed Table 5

and Appendix A.

We found support for H1, i.e., Perceived Ease-of-Use was positively related with

Perceived Usefulness, both for user and non-user, both in Japan and Korea. Moreover, the path

from PEoU explained from 72% to 84% of PU(see values of R-square in Appendix A). However,

we failed to secure a meaningful relationship between PU and ATP for non-user. Thus, H2 was

supported for user but it was rejected for non-user.

Table 5. Summary of Hypothesis test and Significance test between Users and Non-Users

coef. std.err

User 0.545 0.034 15.81 ***

Non-User 0.485 0.039 12.40 ***

User 0.628 0.058 10.81 ***

Non-User 0.515 0.037 13.98 ***

User 0.815 0.171 4.76 ***

Non-User 0.201 0.231 0.87

User 0.829 0.296 2.80 ***

Non-User 0.105 0.163 0.65

User 0.095 0.102 0.93

Non-User 0.325 0.136 2.40 **

User -0.044 0.197 -0.23

Non-User 0.348 0.102 3.42 ***

User 0.903 0.06 14.95 ***

Non-User 0.942 0.052 18.27 ***

User 0.895 0.231 3.87 ***

Non-User 0.96 0.187 5.14 ***

User 0.166 0.064 2.60 ***

Non-User -0.027 0.056 -0.48

User 0.102 0.175 0.58

Non-User 0.218 0.126 1.74

t test Si nificance Test

Usefulness

7/31/2019 10e021

15/22

13

supported for non-user but it was rejected for non-user. This is an interesting finding because it

means that non-user of IPTV significantly consider how he/she can use IPTV easily, while user of

IPTV considers whether IPTV is useful.

H4 was supported. The better attitude toward IPTV people have, the more intention to

use IPTV people have.

H5 was not supporter without user of IPTV in Japan. For all In Korea and for non-user in

Japan, although the relationship has been generally assumed in some related research, our

empirical analysis revealed that PU was not directly related to IUP. When PU was related to IUP

for them, PU was firstly related to ATP which directly related to IUP. Therefore, PU can be related

indirectly to IUP. On the other hand, for user in Japan, PU was directly related to IUP, according

to Table 5 (coef. = .116, p

7/31/2019 10e021

16/22

14

indirectly on the diffusion outcome through Attitude Toward the Product. Again, IPTV is not a

service which potential users with perception of usefulness would jump on. Instead, it is first to

draw attention to the services, then some potential users overcome economical and psychological

barriers existing between attention to actual adoption.

In terms of cross country comparison, we have found similar results for most of paths.

However, there are some differences, particularly in the paths from Perceived Usefulness. One

difference is that a direct influence from Perceived Usefulness to adoption in Japan, while there is

not such relationship in Korea. There are some ways to explain such difference. For example, the

first difference can be explained by the stage of IPTV diffusion in both countries. In Rogers

diffusion model, five types of users are identified by the timing of adoption, i.e., innovators, early

minorities, early majorities, late majorities and laggards (Rogers, 2005). In Japan, since the

diffusion rate of IPTV is only 2.5%, most of current users can be categorized as innovators,

while early minority users also exist in Korea, whose diffusion rate is around 17%. Innovators are

just technology lovers, who tend to jump on new product. Therefore, a positive and significant

path can be find from Perceived Usefulness to adoption directly. Another possible explanation is a

difference in consumer taste between two countries, in a sense that the Japanese put more value

on usefulness as compared to the Korean. This may be related to the difference in another

difference in the path from Perceived Usefulness to Attitude Toward the Product. In Korea, a

positive and significant path can be found only in IPTV users, while it is found only in non IPTV

users in Japan. It should be noted that Perceived Usefulness, as well as easiness of use, is

important for attention even before its subscription for the Japanese, while only perceived

easiness is important for the Koreans.

5. ConclusionsIn this paper, the difference in IPTV diffusion rate between Japan and Korea is

analyzed by estimating structural equation model (SEM), based on Technology Acceptance Model

(TAM) in the literature of innovation diffusion (Davis, 1989; Rogers, 1995). We use the results of

household surveys in Japan, conducted by RIETI and in Korea, by ETRI (Electronics and

Telecommunications Research Institute).

7/31/2019 10e021

17/22

15

Our structural equation model (SEM) analysis reveals that "Ease-of-Use" is an

important factor for non IPTV user to use it. In addition, a direct influence of Perceived

Usefulness to adoption is not observed in general, while its indirect path through attention to

the product is important. Therefore, it is important for IPTV provider to draw attention of

potential users by stressing the easiness of use. In addition, promotion with free trial period may

be an effective marketing tool, to convince non users about its usefulness.

In terms of cross country difference, a direct path from "usefulness" to "adaption" can

be found in Japanese users, while such relationship cannot be found for the Koreans. This can be

explained by difference in the stage of diffusion across countries, in a sense that only innovators

adopt IPTV in Japan, while early minorities is also using it in Korea. Another possible

explanation is cross country difference in consumer taste. We have found a stronger preference in

Perceived Usefulness in Japan, so that appropriate marketing campaign by Japanese providers

may be different from Koreans, by addressing functions and features of IPTV as well as its

easiness of use to potential customers.

This paper sheds new light on the diffusion pattern of new digital convergence services

in Japan and Korea. The results of our estimated structural equation models revealed commonality

and differences in the mechanism of innovation diffusion, which implies some difference in

consumer taste in both countries. However, it is still a long way to go to narrow down more

details in the taste. This is one area for us to investigate further in a next project. In addition, it is

important for us to work on dynamic model, instead of static SEM used in this paper. Another

point to be investigated is a difference in the speed of IPTV diffusion between two countries.

Reference

Agarwal, R. & Karahanna, E. (2000), Time flies when you're having fun: cognitive absorption and beliefs

about information technology usage, MIS Quarterly, 24,4, 665-694.

Ajzen, I., & Fishbein, M. (1980), Understanding attitudes and predicting social behavior. Engelwood Cliffs,

N.J.: Prentice-Hall.

7/31/2019 10e021

18/22

16

Bagozzi. R, Yi. Y (1988), On the evaluation of structural equation models, Journal of the Academy of

Marketing Science 16, pp. 7494.

Cao & Mokhtarian(2005), The Intended and Actual Adoption of Online Purchasing: A Brief Review of Recent

Literature, UC Transportation Center

Davis, F.D. (1989), Perceived usefulness, perceived ease of use, and user acceptance of information

technology, MIS Quarterly, 13 (3), 319-339.

Davis, F.D., Bagozzi, R.P. & Warshaw, P.R. (1989), User acceptance of computer technology: a comparison of

two theoretical models, Management Science, 35 (8), 982-1003.

Fishbein, M., & Ajzen, I. (1975), Belief, attitude, intention and behavior: An introduction to theory and

research. Reading, MA: Addison-Wesley.

Gatignon, H. A., and Robertson, T. S. (1985), A propositional inventory for new diffusion research, Journal of

Consumer Research, Vol. 11, 849867.

Ha. I, Yook. S (2009) The Effects of Media Characteristics on IPTV Adoption, PICMET 2009 Proceedings,

August 2-6, Portland, Oregon USA

Ha. I, Yoo. J, Choi. J, Jong. S, Kim. S, Chin. Y (2009), Adoption of IPTV under the Convergence ofBroadcasting and Telecommunications, Feb. 15-18, 2009 ICACT 2009

Hayduck L. (1987), Structural Equation Modeling with LISREL, Johns Hopkings University Press, Baltimore,

MD.

Hu L, Bentler PM (1999), Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria

versus new alternatives, Struct Equ Modeling 1999;6(1): 155.

Jang Dae-ryeon & Jo Seong-do (2002), A Study on Users of High-Tech Products on Resistance Against

Innovation within Organizations, and Control Effects of Perceived Self-Capability, Korean

Society of Consumer Studies, Vol. 13, No. 3, September 2002, pp.245~262.

Jeffres, L. & Atkin, D.(1996), Predicting Use of Technologies for Consumer and Communication Needs,

Journal of Broadcasting and Electronic Media, 40, 318-330.

Joo, Y.B., Kim, Y.G. (2004), Determinants of corporate adoption of e-marketplace: an innovation theory

perspective, Journal of Purchasing and Supply Management, Vol. 10 No.2, pp.89-101.

7/31/2019 10e021

19/22

17

Jreskog, K. G. (1969), A general approach to confirmatory maximum likelihood factor analysis.

Psychometrika, 34, 183-202.

Jreskog, K. G. (1973), A general method for estimating a linear structural equation system. In Structural

equation models in the social sciences, (A. S. Goldberger & Duncan, Eds.) pp.85-112. Seminar

Press: New York.

Lee, E-J., Lee, J., Schumann, D.W. (2002), The influence of communication source and mode on consumer

adoption of technological innovations, Journal of Consumer Affairs, Vol. 36 No.1, pp.1-27.

Li, Shu-Chu Sarrina (2004), Exploring the Factors Influencing the Adoption of Interactive Cable Television

Services in Taiwan, Journal of Broadcasting and Electronic Media, 48(3), 466-483.

Martin, I. M., and Stewart, D. W.(2001), The Differential Impact of Goal Congruency on Attitudes, Intentions,

and the Transfer of Brand Equity, Journal of Marketing Research, 38, 471-484.

Ministry of Internal Affairs and Communications (2009), State of Cable Television Penetration, Press

Release-Telecom,

http://www.soumu.go.jp/main_sosiki/joho_tsusin/eng/Statistics/pdf/090612_1.pdf

Moore, G. C., & Benbasat, I. (1991), Development of an Instrument to Measure the perceptions of Adopting an

Information Technology Behavior, Information Systems Research, 2(3), 192-222.

Nunnally, Y. J. (1978), Psychometric theory, New York: McGraw Hill.

Parasuraman, A. V., A. Zeithaml, and Berry, L. L.(1985), A Conceptual Model of Service Quality and Its

Implications for Future Research, Journal of Marketing, 49, 41-50.

Price, L. L. & Ridgeway, N. M.(1983), Development of a Scale to Measure Use Innovativeness, in Advances

in consumer Research, 10, 679-684.

Robertson, T. S.; & Gatignon, H(1986), Competitive Effects on Technology Diffusion. Journal of Marketing,

Volume 50, pp. 1-12.

Rogers, E. M. & Shoemaker, Floyd F (1971), Communication of Innovations: A Cross-Cultural Approach (2nd

ed.). New York: The Free Press.

Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York: Free Press.

Rogers, E. M. (1995), The Diffusion of Innovations, 4th ed. New York: Free Press.

7/31/2019 10e021

20/22

18

Scott J. (1994), The measurement of information systems effectiveness: evaluating a measuring instrument,

in: Proceedings of the Fifteenth International Conference on Information Systems, Vancouver, BC,

pp. 111128.

Shin. Y, Jeon. H, Choi. M (2008), Analysis on the Mobil IPTV Adoption and Moderator Effect Using Extended

TAM Model, 4th

International Conference on Networked Computing and Advanced Information

Management

Venkatesh, V. (2000), Determinants of Perceived ease-of-use: Integrating Control, Intrinsic Ease of Use:

Integrating Control, Intrinsic Motivation, and Emotion into the Technology Acceptance Model,

Information System Research, 11(4), 342-365.

Yoo. J, Choi. J, Kim. S, Yang. J, Jong. S (2009), Factor Affectiong the Switching Intention of IPTV

Subscribers, Feb. 15-18, 2009 ICACT 2009

7/31/2019 10e021

21/22

19

Appendix A

ATPR2=0.439

R2=0.177

ITPR2=0.630

R2=0.690

PEoU

PU

R2=0.767

R2=0.723

***0.166-0.027

***0.903

***0.9420.095

**0.325

***0.8150.201

***0.545

***0.485

Above: user, Below: non-user, ***:1% significant, **: 5% significant

Figure A1. Results of SEM in Japan

ATPR2=0.505

R2=0.674

IUPR2=0.523

R2=0.487

PEoU

PU

R2=0.814

R2=0.747

0.1020.218

***0.895

***0.960-0.044

**0.348

***0.8290.105

***0.628

***0.515

Above: user, Below: non-user, ***:1% significant, **: 5% significant

Figure A2. Results of SEM in Korea

7/31/2019 10e021

22/22

20

Appendix B

Table B. Summary of Hypothesis test and Significance test between Japan and Korea

coef. std.err

Japan 0.545 0.034 15.81 ***

Korea 0.628 0.058 10.81 ***

Japan 0.485 0.039 12.40 ***

Korea 0.515 0.037 13.98 ***

Ja an 0.815 0.171 4.76 ***

Korea 0.829 0.296 2.80 ***

Japan 0.201 0.231 0.87

Korea 0.105 0.163 0.65

Ja an 0.095 0.102 0.93

Korea -0.044 0.197 -0.23

Japan 0.325 0.136 2.40 **

Korea 0.348 0.102 3.42 ***

Japan 0.903 0.06 14.95 ***

Korea 0.895 0.231 3.87 ***

Japan 0.942 0.052 18.27 ***

Korea 0.96 0.187 5.14 ***

Japan 0.166 0.064 2.60 ***

Korea 0.102 0.175 0.58

Ja an -0.027 0.056 -0.48

Korea 0.218 0.126 1.74

t test Si nificance Test

Usefulness

Documents

10e021